Deep Non-rigid Structure-from-Motion Revisited: Canonicalization and Sequence Modeling
This work improves 3D reconstruction from 2D sequences for computer vision applications, representing an incremental advancement in deep NRSfM methods.
The paper tackles the problem of Non-Rigid Structure-from-Motion (NRSfM) by addressing limitations in handling sequence properties and motion ambiguity, achieving a more optimal reconstruction pipeline through canonicalization and sequence modeling methods.
Non-Rigid Structure-from-Motion (NRSfM) is a classic 3D vision problem, where a 2D sequence is taken as input to estimate the corresponding 3D sequence. Recently, the deep neural networks have greatly advanced the task of NRSfM. However, existing deep NRSfM methods still have limitations in handling the inherent sequence property and motion ambiguity associated with the NRSfM problem. In this paper, we revisit deep NRSfM from two perspectives to address the limitations of current deep NRSfM methods : (1) canonicalization and (2) sequence modeling. We propose an easy-to-implement per-sequence canonicalization method as opposed to the previous per-dataset canonicalization approaches. With this in mind, we propose a sequence modeling method that combines temporal information and subspace constraint. As a result, we have achieved a more optimal NRSfM reconstruction pipeline compared to previous efforts. The effectiveness of our method is verified by testing the sequence-to-sequence deep NRSfM pipeline with corresponding regularization modules on several commonly used datasets.